Plants > Biotechnology / PNTs > Perspectives on Novelty 

Perspectives on Novelty and the PNT Approach

Proceedings from Recent March 2004 Workshops

October 27, 2004

Table of Contents

1. Forward

This document summarizes recent activities of the Canadian Food Inspection Agency (CFIA) and Health Canada to help clarify the application of the regulatory trigger for plants with novel traits (PNTs), novel livestock feeds and novel foods. It begins with an introduction about the origin and history of PNT, novel feed and novel food regulations. This is followed by proceedings from two workshops attended by government regulators, plant breeders and product developers to discuss the use of novelty as the regulatory trigger. The document closes with a summary of recent actions that the CFIA and Health Canada have taken to address issues pertaining to novelty, and includes a brief project for future activities.

As with any regulatory approach, the use of novelty as the regulatory trigger has advantages and disadvantages. It has been said that this approach impacts broader issues such as the market acceptance of new products, public perceptions of product safety, and the Canadian agricultural sector’s interactions with the rest of the world. However, the focus of the two novelty workshops and this summary document is narrow. The question at hand is how to clarify the interpretation of novelty for plant breeders, product developers and government regulators. Issues beyond this question, while important, are outside of the CFIA’s and Health Canada’s regulatory authority and consequently, are not addressed in this report.

All participants involved in the discussion to date should be commended for their willingness to tackle a complex issue and to engage in challenging discussions. There have been many useful suggestions for a path forward that will help regulators continue to work with the plant breeding community to clarify the determination of novelty.

2. Introduction

The use of genetic engineering as a tool for plant breeders, advances in scientific knowledge and, growing public consciousness about environmental and food safety issues was the driving force behind novel plant varieties receiving pre-market assessments of their environmental and livestock feed safety for the first time in the mid-1990s. At the same time, the presence of novel traits in food and livestock feed became a trigger for pre-market safety assessments. This represented a significant shift in philosophy about the regulation governing the introduction of new plant varieties in Canada. Indeed, new varieties of familiar or conventional crops had been bred, sold, planted, and consumed in Canada for decades without such pre-market safety assessments.

Several events and factors led to this shift in philosophy. For example, in 1986, the Organisation for Economic Co-operation and Development (OECD) published Recombinant DNA Safety Considerations, which provided a set of requirements for genetically engineered products. It proved to be an important document, used by many nations to develop their regulatory framework. The follow-up document, Safety Considerations for Biotechnology (OECD, 1992), outlined safety assessment requirements, especially related to environmental issues. Interestingly, neither document prescribed what biotechnology was; therefore, each country defined the concept independently leading to the varying definitions found worldwide today. Following publication of the two OECD documents, several countries, including Australia, Japan, the United Kingdom, the United States and Canada, developed and implemented regulatory requirements.

Additionally, the Canadian Environmental Protection Act (CEPA) was enacted in 1988 with a requirement for regular review. CEPA established a requirement for any person wanting to import, manufacture or sell a "new substance" to notify the appropriate Canadian regulatory authority so the product could be evaluated for potential effects on the environment and human health.

It was in 1988 that a product-based, rather than process-based, trigger for safety assessments of novel plants was recommended to the Government of Canada. This recommendation emerged from a key Canadian Agri-Food Research Council (CARC) workshop of respected Canadian scientists on the Regulation of Agricultural Products of Biotechnology. CARC is a not-for-profit consortium of researchers from industry, academia, and federal and provincial governments. It reflected a discussion that all novel plant varieties may potentially pose risks to the environment, food or livestock feed, regardless of the genetic modification method used to produce them. This direction from CARC was pivotal in assisting the government in the development of a regulatory framework for novel products.

With regard to novel food regulation, the concept of pre-market clearance for novel foods was first discussed in 1992. The globalization of the food supply, the demand for more food sources globally, and the rapid advances in food science and technology have resulted in the introduction of foods not previously available in the Canadian marketplace. In response to the introduction of these novel foods, Health Canada enacted the Novel Foods Regulation under the Food and Drugs Act, which requires manufacturers and importers of novel foods to notify the Department before the sale of the product in Canada. Based on the definition of a novel food, foods derived from PNTs could trigger the notification requirement of the Novel Foods Regulation.

Later in 2001, the Seeds Act and the Feeds Act were scheduled under CEPA, an action that formally transferred certain regulatory responsibilities to the CFIA.  Under CEPA, new products were referred to as new substances. In 2001  a policy decision was made to continue using the terms plants with novel traits (PNTs), and novel feeds instead of new substances.

Recently, the developers of new plant varieties and other stakeholders have asked the CFIA and Health Canada to clarify the use of novelty as their regulatory trigger. The latest discussion on novelty began at the consultations related to the revision of directives: Dir94-08: Assessment Criteria for Determining Environmental Safety of Plants with Novel Traits, Dir95-03: Guidelines for the Assessment of Livestock Feed from Plants with Novel Traits, and Guidelines for the Safety Assessment of Novel Foods. This process began with a multi-stakeholder consultation in 2002, and posting draft revised guidelines for comment in 2003. Many comments received in both of these forums pertained to clarifying the use of novelty as the regulatory trigger.

The "PNT approach" was also discussed during the 2003/2004 Seed Sector Review. In the proceedings it was noted that, while most participants thought that the PNT approach was a good one, there was also a lack of a clear understanding about what ‘novel’ means in a regulatory context.

The PBO, the Feed Section of CFIA and the Novel Foods Section of Health Canada recognized, based on discussions about the concept of novelty mentioned above, that there was a need to meet with stakeholders specifically to discuss this topic. Consequently, two meetings were organized in the spring of 2004 to discuss the use of novelty as the regulatory trigger. The proceedings from these meetings follow.

3. Saskatoon Meeting Proceedings

The Ottawa novelty workshop was informed by an earlier meeting between CFIA regulators and plant breeders from industry, academia and AAFC, held in Saskatoon on March 19, 2004. The workshop was organized to allow greater engagement of interested parties who are based in Western Canada on the application of novelty as the regulatory trigger for PNTs and novel livestock feeds.

3A. Participant List

3B. Agenda

3C. The Presentations

The History of the Novelty Concept was presented by Grant Watson, CFIA. The presentation is included in Appendix 1, on page 30.

Novelty as the Regulatory Trigger for Feed was presented by Lynne Underhill, Feed Section, CFIA. The presentation is included in Appendix 1, on page 34.

Novelty as the Regulatory Trigger for Plants was presented by Phil Macdonald, Plant Biosafety Office, CFIA. The presentation is included in Appendix 1, on page 36.

3D. Commentary from the Discussion¹

Comments and questions from a plenary discussion involving all participants were recorded. This information has been presented in two sections. First are questions posed to regulators, with their responses, and second are the take home messages participants were asked to provide at the close of the discussion.

Questions for the Regulators, and their responses:

Why regulate based on novelty, rather than risk?

• Novelty is only the trigger for a safety assessment. We can’t know what’s risky until the assessment is done. Once an assessment is triggered by a novelty determination, the basic data requirements are outlined in Directives 94-08 and 95-03. However, there is flexibility to allow product developers to make scientific arguments about why certain safety criteria or tests might not apply to a certain PNT.

The CFIA asserts that their regulations are not processed-based, but wouldn’t any rDNA plant trigger regulation?

• It may be possible that an rDNA product will not be considered novel, based on its characteristics. For example, the CFIA’s policy on retransformations states that they may not need to be reassessed, however notification is required.

Why would seed coat colour trigger novelty? The substances that create a new seed coat colour were always present in the seed.

• The definition of "novel" in a regulatory context doesn’t require the presence of an entirely new compound. If a trait is expressed significantly outside of its known range, it would be considered novel.

Plant Breeders do a lot of work in the field before plants reach the Variety Registration Office (VRO), how is the environment protected at this stage?

• A new variety should be identified as a PNT before it reaches the VRO. As soon as a novel trait has been identified, it should be recognized that the plant is regulated and the plant should be grown in the PNT field trial system.

The only way to determine if a new variety is novel is for plant breeders to grow it out in the field. How can all of our research plots be regulated?

• As regulators, we’re not trying to regulate all plant breeding or prevent plant breeders from making selections in the field. The majority of new plant varieties are not going to be novel (in the regulatory context). PNT regulation is not triggered until a novel trait is actually selected. At that time, a plant breeder should contact the PBO about conducting PNT confined field trials.

How do regulators define environment?

• Environment is defined in the regulations as: Components of the earth (including air, land, water, all layers of the atmosphere, all organic and inorganic matter and living organisms) and all interacting natural systems that include components referred to above (Canadian Environmental Protection Act, Section 3). This includes the natural and managed ecosystems, which includes agricultural ecosystems.

Until I know if a trait I’ve selected is stable, how can I know if it’s novel?

• When a plant breeder selects a trait that is novel, then the PNT regulatory trigger has been tripped. At that point, the PNT should be grown in a regulated confined field trial. It doesn’t matter if the trait is stable or not. Field trials are intended to provide product developers with a mechanism to study the PNT, and to collect data to fulfill the requirements of an unconfined release safety assessment. One of these data requirements is a demonstration of trait stability. Thus, trait stability is typically studied in confined field trials, not prior to determining if a PNT has been selected.

Take-Home Messages from the Plant Breeders:

• A clear definition of novelty is needed so that all parties who use it understand its meaning in the same way.
• There is a need to have clear parameters around the definition of novelty. PBO should take the lead in making sure there are consistent parameters.
• Crop committees/registration recommending committees should come up with guidelines for PNTs.
• The existing system has been working. An advantage is that plant breeders have the freedom to assess their own plants, and to make their own determinations about what’s novel.
• Improvement is needed to the transparency of the regulatory system, especially with regard to retransformation events.
• Canada doesn’t operate in isolation from the rest of the world. We should regulate GMOs, not PNTs, to reflect the global environment with respect to science and economics.
• It’s a concern that Canada is neither in step with European scientist thinking nor on side with the American approach.
• "Novelty" should be replaced with a new word that has one clear meaning and a positive connotation.
• The focus today has been on triggers, but we need to spend as much time on what we do with it once that trigger is tripped - 90 per cent of products that are deemed to be novel will pose no risk.
• Falling into the PNT category raises concerns that our plants will be perceived to be risky when they aren’t.
• It is harder to be competitive with the rest of the world, when we have to contend with the cost and time restrictions associated with PNTs.
• Plant breeders have not been heard, what they want in the definition has not been incorporated in the past.
• The PNT definition is ludicrous and wrong, but we have to live within the system we have. We need a more simplistic means to get more quickly through the system if the risk of a new product is low. We need more guidance about what has to be done when a new variety has a novel trait.
• If this approach is truly science-based, it doesn’t work to wait until a novel trait is selected by a plant breeder, because the trait was present in the breeding population, already out in the environment, when it was identified by the breeder.
• Everyone needs to know what the rules are. I now have a clear understanding that until I start selecting for a novel trait, I’m not regulated. Now we should be able to start that process before we get to VRO.

Take-Home Messages from the Regulators:

• We understand that plant breeders are trying to do good things for Canada and we’re not unsympathetic to the fact that the PNT approach has an impact on your work.
• There are benefits for plant breeders and product developers to having a product-based regulatory system in place. For example, this approach treats all breeding methods equally. However, recognizing that other countries use process-based systems, we indicate on the CFIA web site whether or not PNTs are also products of rDNA technology.
• We’re trying to take the regulatory system we have and make it work better. We don’t want to regulate all plant breeding. We want to set the regulatory bar high enough to continue to have a workable, science-based system. To do that, we need more involvement from the expert community.
• The PNT approach has come up when we discuss the scope of variety registration, and in the seed sector review process. Environmental concerns such as biodiversity and invasive species are on the minds of the public. The issue of environmental safety of PNTs is not going to go away, and there is a significant community of interest in how we regulate.
• In the future, the degree to which our existing system is strong enough to address new challenges will be important. We hope to have your help in making it work better.

4. Ottawa Workshop Proceedings

Plant breeders from industry, academia and Agriculture and Agri-Food Canada (AAFC) research centres met with regulators from the CFIA and Health Canada at a workshop held in Ottawa on March 29 and 30, 2004. The workshop was organized to discuss the application of novelty as the regulatory trigger for plants with novel traits (PNTs), novel livestock feeds and novel foods. The influence of plant breeding methods on unanticipated effects of plant breeding was also discussed.

4A. Participant List

4B. Agenda

4C. Presentation Summaries

William Yan, Health Canada

Dr. William Yan, on behalf of the Food Directorate at Health Canada, as well as both the Plant Biosafety Office and the Feed Section of the CFIA, welcomed participants to the workshop. Key points were:

• This meeting was conceived from a need government regulators have, and that plant breeders have also identified, to improve our ability to decide when a new variety is new enough to come within the regulatory statutes governing plants with novel traits, novel foods and novel feeds. Terms like "substantially equivalent", "familiarity" and "within the anticipated range" in our respective definitions of novelty could benefit from further clarification.
• In preparing the agenda, we also considered that there could be better awareness among the various non-government stakeholders concerning what the government perspective is and why we have the regulations we do, wherein the trigger for regulation is "novelty". And so this meeting is an opportunity to address some of those gaps.
• At this meeting, CFIA and Health Canada are not expecting to come to a set of conclusions or a set of fully formed guidelines. What we intend to do is take what we learn here and, in subsequent meetings among the various offices involved in regulating plants with novel traits and their products, set in motion a process for drafting appropriate guidance materials.
• Dr. Yan concluded by wishing the group a fruitful series of discussions and success in achieving greater understanding from all sides of the question that will provide a solid basis for moving to next steps.

Phil Macdonald, CFIA

Plants with Novel Traits - Novelty as a Regulatory Trigger

Phil Macdonald spoke about the PNT regulatory approach from CFIA’s point of view. Key points were:

• The product-based trigger for regulation in Canada is broader than the process-based system currently used in the rest of the world, but provides equal regulatory oversight for all Canadian developers of new plant varieties
• This approach was developed following extensive multi-stakeholder consultations, conducted by AAFC, and was supported by the agriculture sector.
• This product-based approach is confirmed in the first version of the Directive on the environmental assessment criteria for plants with novel traits published in 1994 (Dir94-08) and on the livestock feed assessment criteria for novel feeds published in 1995 (Dir5-03).
• These directives (Dir94-08 and Dir95-03) clearly state that the scope of regulation is greater than just plants derived through recombinant DNA technology.
• The responsibility for determining novelty rests with the developer of the new product.
• A determination of novelty does not necessarily imply that the novel trait poses a risk; it is only a trigger for a safety assessment.
• The CFIA takes a case-by-case approach to the assessment of PNTs and novel feeds. Product developers may have less extensive data requirements to meet, if a valid scientific argument is provided to demonstrate that there is no reasonable expectation of significant risk to the environment.
• We appreciate your concerns regarding potential trade issues resulting from confusion between transgenic and non-transgenic PNTs. The table of approved PNTs posted on the PBO website distinguishes between Living Modified Organisms (LMOs) and other PNTs. We will be happy to discuss other ways that we can clearly distinguish PNTs that are not regulated outside Canada, for the sake of both export and domestic markets.
• The CFIA is always willing to aid a developer in a determination of novelty and continues to work with experts to make sure that we are capturing truly novel products while allowing the plant breeding community to continue to develop new varieties.
• The CFIA has embarked on a major revision to its directives on the environmental assessment criteria for novel plants and on the livestock feed assessment criteria for novel feeds. The revised directives include a section to provide developers more guidance on determining novelty.
• We look forward to continuing to work with the industry and developers of new varieties to further enhance this section of the directives.

Nora Lee, Health Canada

Novel Foods: Clarifying Regulatory Objectives

Nora Lee spoke about the development of Health Canada’s novel food regulatory approach. Key points were:

• The concept of pre-market clearance for novel foods was first mentioned in Information Letter No. 806, in 1992, on Novel Foods and Novel Processes.
• These regulations for novel foods and novel food processes were developed to ensure that the ever-expanding and innovative developments in food science and processing do not adversely affect the safety of the existing food supply, while allowing continued development in these areas. In addition, these changes made the Food and Drug Regulations compatible with the regulatory intentions of the Canadian Environmental Protection Act (CEPA).
• The definition of ‘genetically modify’ in the Food and Drug Regulations means to change the heritable traits of a plant, animal or microorganism by means of intentional manipulation.
• The definition of novel food in the Food and Drug Regulations includes food that is derived from a plant, animal or microorganism that has been genetically modified such that:

  (i) the plant, animal or microorganism exhibits characteristics that were not previously observed in that plant, animal or microorganism,
  (ii) the plant, animal or microorganism no longer exhibits characteristics that were previously observed in that plant, animal or microorganism, or
  (iii) one or more characteristics of the plant, animal or microorganism no longer fall within the anticipated range for that plant, animal or microorganism.

• In 1995 changes were made to clarify that notifications would only be required for new processes that are truly new and cause substantial changes in the food.
• Plants with traits that have been either introduced or removed by genetic modification are automatically considered sufficiently changed to warrant a notification of foods derived from them.
• Where the change is a matter of degree, only where it is large enough to bring the trait outside of the anticipated range should foods from those plants be considered novel.

Sher Bushell, CFIA

Variety Registration

Sher Bushell spoke about continuity between PNT regulation and variety registration. Key points were:

• When applying for variety registration, applicants must indicate if the variety contains a novel trait.
• The onus of classifying a crop variety as containing a novel trait rests solely on the applicant. However, there are cases where the candidate variety was not deemed to be novel by the applicant, but has been identified by the Variety Registration Office as a potential PNT and vice versa.
• When the novelty status of the variety is questionable, the Variety Registration Office may request data and a scientific rationale from the applicant to support their determination and/or seek advice from Health Canada, Feed Section and the Plant Biosafety Office.

Elizabeth Lee, University of Guelph

Plant Breeding – philosophical perspective

Elizabeth Lee spoke about the complexities of plant breeding and her perspectives as a white corn breeder. Key points were:

• The basics of plant breeding have not changed for centuries. However, our understanding of what is going on during the process of plant breeding has greatly expanded, and the tools that plant breeders can use have changed over time.
• Plant breeding is NOT introducing transgenes into plants. This is just another tool for plant breeders to use.
• Phenotype = Genotype + Environment + (Genotype x Environment) + error.

  For example:

  • Phenotype: Pink kernel striping (PKS) in white corn
  • Genotype: caused by an allele of the Pl1 gene
  • Environment: PKS is only observed when lines are grown in Canada
  • G x E: PKS-pl1 allele + right environment = PKS
  • Error: Sometimes, for whatever reasons, the PKS-causing alleles do not cause PKS.
• Some definitions presented include:
  • Genetic variation: multiple alternative forms (alleles) of each or many genes.
  • Germplasm: Varieties, landraces, populations, related taxa, and wild relatives of the plant species.
  • Genome: Nuclear DNA + Mitochondrial DNA + Plastid DNA
  • Nuclear DNA: genic DNA + non-genic DNA
• The three dogmas of creating genetic variation (recent science has challenged all three):

  #1. In plant breeding, most new genetic variation is due to mutations.

  #2. DNA sequence = phenotype

  #3. All alleles in an individual have allelic counterparts in a second individual.

• In the past, recombination was thought to occur between non-genic regions. The reality is that recombination mainly occurs between genic regions. The genome is under constant change.
• Causes of genetic fluidity are epigenetics, unequivalent genetic content, and to a lesser extent, rapid mutational events.
• Things to think about:
  • The genome is a weird, wild, wonderfully interesting place. We do not understand how things work at any level.
  • Novel genetic variation is constantly arising in Canadian breeding nurseries every year.
  • Method of creating "novel" genetic variation needs to be part of the definition.
  • If it exists within that Taxa somewhere in the world then it should NOT be considered "novel".
  • P = G + E + G x E + error. You could have "novel G", but not the correct E in which to see a "novel P". What is "novel", the G or the P?

Brian Miki, AAFC

The Stability of the Transcription of Transgenic Plants

Brian Miki presented results from his work on examining insertional effects of transgenes. Key points were:

• This research was initiated to address questions about the unintended effects of transgene insertion, specifically:
  • What is the actual number of genes with altered expression levels resulting from transgene insertions?
  • Are the number of genes with altered expression very different from those induced by common environmental signals?
  • Does the insertion of a transgene alter the kinds of genes expressed?
  • Does the presence of a transgene alter the plant responses to environmental signals under stress?
• The numbers and types of up-modulated and down-modulated genes were compared in Arabidopsis in response to a number of abiotic stresses including cold, drought and salt. Similarly, the number and type of up-modulated and down-modulated genes was compared in Arabidopsis in response to the insertion of different transgenes, including NPTII, GUS and BBM.
• Preliminary results indicate that the non-target effect of transgene insertions on the patterns of global gene expression is very small, and changes in global gene expression associated with adaptation to environmental factors is orders of magnitude greater.
• Pleiotropic effects caused by transgene insertion could be easily masked by the high level of naturally-occurring variation.
• In general, the extent of phenotypic changes correlated with the extent of changes in global gene expression.

Alan McHughen, University of California Riverside

Unintended effects from non-rDNA plant breeding

Alan McHughen spoke about unintended effects that can occur during conventional plant breeding. Key points were:

• Unexpected effects can occur with any method of breeding. Some examples include:
  • potato with high solanine (several cases)
  • celery with excessive psoralen
  • tomatoes with high tomatine
  • kiwifruit with allergens
  • squash with high curcurbitacin E
• There are likely many such examples which go unknown, unnoticed and undocumented.
• An unexpected result may not, by itself, invoke hazard.
• Inter-species gene transfer is common and usually innocuous, occurring for example in wheat and various Brassicas. Examples of intentional interspecies crosses include tomato with over 100 L. peruvianum genes, wheat chromosome engineering (esp. rye), and potato hybrids (e.g. S. brevidens).
• Method of breeding is immaterial to risk analysis, which demands a consideration of:
  • the host species,
  • the novel trait, and
  • the environment in which it will be released.

4D. Commentary from the Discussions

Participants were organized into five groups for discussion. Not all groups discussed the same questions due to time constraints and the diversity of experience and interests within the groups. The following is a summary of the groups’ responses to questions the CFIA and HC posed, or that participants identified during plenary discussions.

Q1.

What are current practices used by plant breeders to characterize their plants? Do plant breeders routinely monitor for unintended effects?

A plenary discussion on this question determined that the answer is highly crop-specific. However, there are certain generalizations that can be made. Crops that are subject to variety registration are evaluated by recommending committees or other bodies. Typically, agronomic quality traits and morphological traits are evaluated, which are potentially good indicators of unintended effects that may occur during the breeding process.

With regard to environmental safety, food safety and feed safety, there are some routine checks performed by plant breeders on certain crops. For example, the level of glycoalkaloids in potato tubers is routinely determined for advanced lines in modern breeding programs. Oil, protein and glucan levels are routinely tested in oat breeding. However, there is not a standard approach to testing for antinutrients or toxins in the majority of cropkinds.

Participants said that this question would be best answered on a crop-by-crop basis through consultation with relevant experts such as breeding committees, recommending committees, the Canadian Seed Trade Association, the Canadian Horticultural Council, and industry associations.

Q2.

How might we define "trait"?

One group discussed that a trait is the combination of a phenotype and the biological basis for that phenotype. The same phenotype can be the result of different biological mechanisms. For instance, resistance to European corn borer (phenotype) in corn can be due to expression of Bt proteins, proteinase inhibitors, or synthesis of DIMBOA (in other words, different biological bases).

Another group discussed that a trait is the intended phenotype for which a plant breeder selects. This may include a novel protein, new metabolites or pathways. This group noted that, in many cases in conventional breeding, varieties are developed that express new phenotypes for which the underlying genotype or mechanism of action is unknown or not well characterized. In these cases, any regulatory activities that could be based on the mechanism of action for the trait would be difficult to perform. This group considered that the determination of novelty should include consideration of the end use of the phenotype and be determined on a case-by-case basis. For example, it was suggested that altering the nutritional composition of a plant may warrant only a novel food and feed safety assessment but not a complete environmental safety assessment.

Q3.

What should be the reference base for determining novelty?

One group discussed that, with respect to environmental safety assessments, the germplasm base must be defined within the species and grown in Canada. But, for foods and possibly feed safety assessments, a broader reference base for germplasm or food/feed composition databases could be acceptable since the food we eat regularly comes from beyond our borders.

Another group discussed that novelty should be determined by comparing the intended phenotype with a similar existing characteristic in the plant species in Canada. This group also discussed the geographical range used when determining novelty. Plant breeders at this table suggested expanding the trigger to allow data from outside Canada when determining novelty, towards making the regulatory trigger manageable. Regulators at this table disagreed, because familiarity of a crop is based on what is grown in the Canadian environment, and because trait/environment interactions may be quite different within Canada than outside Canada.

Q4.

What is a stable population?

One group discussed that a stable population is a sizable germplasm pool with predictable trait frequency recurring from year to year and across environments.

Q5.

How should the anticipated range of a trait be determined?

Several groups reported similar ideas in response to this question. Responses from different groups have been interwoven in the following summary:

How can anticipated range be defined?

The anticipated range of specific components expressed in each crop should come from a wide range of different cultivars grown under a variety of environmental conditions. The magnitude of the difference in the parameters measured in the new plant compared with the anticipated range of similar existing parameters in currently grown cultivars could form the basis of determining novelty.

Some participants felt that the anticipated range should include data generated in the Canadian environment, or, with regard to food and feed, ranges appropriate for "history of use in Canada." However, other participants felt that broader geographical ranges should be considered.

Some participants felt that, in some cases such as altered oil composition, data from one crop species could be applied to other crop species in novelty determination.

Some participants felt that certain types of traits could be exempted from the novelty discussion. For example, it was suggested that changes in agronomic traits such as yield would not likely result in novelty and that a broad general exemption could be made. This was based on the typically incremental changes that occur in agronomic traits in breeding programs, and the tendency for agronomic improvements to reduce the fitness or life history traits of a crop. However, a caveat was included that novelty should still be considered if a highly unusual change resulted from a breeding program. Other participants did not support the idea of a broad general exemption for certain agronomic traits.

Start with the literature

Trait ranges can be based on literature values, however information may not be current and complete for all crop/trait/environment combinations. Literature values may be supplemented by data from within the breeder’s own program, i.e. by comparing the selected variety to a range of common commercial cultivars being used as controls in the breeding program. Another potential source of data is from previous variety registration trials or other performance trials for the crop.

Document ‘what’s known’

Rather than trying to characterize "What is novel?" or "How novel is novel?" it might be more practical to document what we already know. This could be accomplished either by adding information to existing biology documents used in the PBO, by drawing on OECD consensus documents, or by generating new documents that specifically describe anticipated ranges for selected traits on a crop-specific basis. Such documents would provide plant breeders with an early warning that their particular plant may be considered novel, and they could be used as the basis for determining novelty, i.e. for deciding if a selected trait is outside the range for a specific plant.

PBO crop biology documents could be substantially expanded to include key traits, along with data where available. Traits would be relevant to environment, food, and feed perspectives, in addition to the basic biology of the crop species,

Proposed documents could be written, developed, reviewed, and/or updated by the scientific expert community. This community could be formed from existing expert committees, plant breeders, nutritionists, toxicologists, plant pathologists etc.

The documents could be transparent/publicly available on a web site such as CFIA’s, and regularly updated.

Use expert committees

In order to determine if a trait is substantially equivalent to a similar existing trait in that plant species, breeders require additional information and guidance to determine if the new trait expressed in the plant is present in cultivated populations in Canada, in the food supply or in imported crops.

This could take the form of a committee of experts from the plant breeding community, government regulators and industry. This committee should be crop-specific, and would tap into existing expertise. This approach would help make sure that the government is working with the most current information about a specific crop, since ranges of specific traits are moving targets.

However, the use of such committees raised concerns about confidentiality and competition among some participants. It was suggested that consultation with such a committee could occur on a voluntary basis.

Q6.

Do regulators need to do more to make the plant breeding community aware of these regulations? Where and how would our efforts be most effective?

This question was discussed in plenary. The group indicated that regulators should do more to communicate about the regulations. Giving talks at scientific conferences, Variety Registration Recommending Committee meetings, and at industry meetings such as the Seed Sector review and value chain round tables were mentioned.

4E. Questions about Novelty²

Questions that arose during the first half of the workshop were recorded. Participants were later asked to select the five questions which they viewed as the most important to furthering the discussion. The complete list of questions that arose is provided below. The number of times a question was selected for discussion is indicated in brackets.

1. How can we define the reference population or comparator? (13)
2. How does the use of the plant factor in to the regulatory trigger? For example, what if a conventional plant variety was started to be used for phytoremediation?
3. Is novelty limited to heritable traits? What about transient expression?
4. How long must a trait have been present before it is no longer novel? (3)
5. What about novel species?
6. How can we better define triggers for non-rDNA PNTs? (19)
7. What geographical range can be used to inform Canadian safety assessments?
8. How should substantial equivalence be defined/used? (4)
9. How can the lack of genomic data for minor crops be addressed? Can data for one crop be used for another? How? In what circumstances might this be allowed? (6)
10. How do plant breeders know if unintended effects have occurred during the plant breeding process? (3)
11. What are strategies to assess unintended effects that may occur during the plant breeding process? (12)
12. When directives refer to a ‘stable population’, how is this term defined? (2)
13. When directives refer to a ‘cultivated species’, how is this term defined? (3)
14. How can the ‘anticipated range’ of a trait be determined? (22)
15. Can recombination result in changes to the nutritional composition or (toxic) protein profile of a new plan variety? (2)
16. How can the ‘collective wisdom’ of the plant breeding community be accessed by regulators? (2)
17. To what extent should plant breeders be expected to look for intended or unintended novel traits? (2)
18. How should the regulatory trigger apply to qualitative vs. quantitative traits? For example, changes in some traits occur slowly, due to the cumulative impact of several genes. (2)
19. If fruit was produced on a transgenic tree, but the fruit itself did not contain a transgene, would this be considered a novel food?
20. If a trait is not found in a stable population of a specific species, but is found in other crops/species, could a case be made that the trait is not novel? (3)
21. In 1988, several ‘categories of concern’ were identified, including novel herbicide tolerances, novel pesticide properties, novel stress tolerances and novel compositional changes. Do these still apply? Are there others? (7)
22. What do consumers need to be confident in the regulatory system? (1)
23. What happens if a novel trait occurs or is found post-commercialization?
24. What is the potential impact of environment on a novel trait?
25. How will the import of food from non-rDNA PNTs be addressed?
26. Given the degree of variation in all plant breeding, how do we choose where to place the regulatory bar, and what to regulate? (12)

5. Recent Actions on Novelty and the Path Forward

Recent actions at CFIA and Health Canada regarding the novelty issue include the revision of directives and the publication of a brochure about novel feed. As an initial next step to provide the plant breeding community with additional guidance on novelty, the Plant Biosafety Office is planning to coordinate the preparation of a guidance document. This project is outlined below, following a brief description of other recent activities.

5A) Revised directives – CFIA:

The Plant Biosafety Office and Feed Section, CFIA, have worked to clarify the definition of novelty in the revised draft of directive Dir94-08: Assessment Criteria for Determining Environmental Safety of Plants with Novel Traits and Dir95-03: Guidelines for the Assessment of Novel Feed: Plant Sources. The final version of the directives are published on the CFIA website. The final version contains a clearer, more detailed description of the regulatory trigger and some revised definitions associated with novelty.

5B) Revised Guidelines – Health Canada:

Health Canada has completed a public consultation on the revised Guidelines for the Safety Assessment of Novel Foods Derived from Plants and Microorganisms. These guidelines establish the safety assessment criteria that assist developers in collecting the information required to demonstrate the safety of novel foods. The guidelines were revised based on the experience gained in the Department’s evaluation of these products over the past 8 years. In addition, the promulgation of the Novel Foods Regulation in 1999 required that these guidelines be updated to address the legislated responsibility for pre-market assessment of novel foods outlined in Division B.28 of the Food and Drugs Act. Text has been added to the revised guidelines to assist developers in interpreting the novel food definition. In addition, a new standard operating procedure has been developed by the CFIA and Health Canada to improve the coordination of determining novelty between the two regulatory bodies.

5C) Publication of a Novel Feeds Brochure

In March 2004, the CFIA’s Feed Section published a brochure Regulation of Novel Feeds, which includes information about the regulation of novel feeds derived from plants, animals and microbes. The brochure contains a section titled "How do I know if my feed is novel." The brochure can be found at: http://www.inspection.gc.ca/english/anima/feebet/bio/regulatione.shtml.

5D) Path Forward - A Novelty Guidance Document

 The PBO will consult with interested stakeholders in March 2005 about the scope, purpose, and management of a document designed to provide useful advice and recommendations to assist in the determination of whether or not a new plant variety is a PNT/novel feed.

Although the contents of the document will evolve to suit the needs of the plant breeding community, elements of the document may include:

• detailed examples of novelty determinations for specific crops and traits (may be real or hypothetical).
• detailed examples of the ranges of certain traits, and typical changes in certain traits, for specific crops.
• information about how specific crops are characterized before new varieties or lines are commercialized.
• suggestions for when and how to communicate with regulators about new plants that may be PNTs/novel feeds.

The PBO has surveyed workshop participants and other experts who have expressed interest in novelty determination, to gauge their level of support for developing a guidance document and their willingness to participate. CFIA is presently working with Health Canada to develop a skeleton for the guidance document that will be presented at the March consultation.

¹ As Health Canada was not represented at the meeting, please note that the scope of the discussions included environmental safety and feed safety, but not food safety.

² Some of the questions are about issues that are outside of the authority and mandate of the regulatory offices at CFIA and Health Canada to address or change.